Combining imaging techniques and CFD to model lung deposition in various age classes of the paediatric population

Combining imaging techniques and CFD to model lung deposition in various age classes of the paediatric population

Contents: Motivation: Why modelling the paediatric dose to lung via in-vitro and in-silico methods? Experiences with adult models Exploring the paediatric anatomy and physiology Why to use paediatric models at an early stage of product development? Challenges in the lab Summary & Outlook 2

The individual dose to lung is the result of a complex interaction between patient, inhalation device and formulation! Age Inhalation technique Excipients Physicochemical properties Patient Formulation Anatomy & Morphology Handling MMAD, PSD, FPD Manufacturing process Air flow resistance Device Spray velocity, Nozzle / Mouthpiece geometry Dispersing mechanism 3

A big challenge for inhaled medications: The 90 bend of human mouth-throat Scintigraphic pictures of particle deposition of 99MTc-labelled Tiotropium (T Meyer et al. / 2003) 4

A further big challenge for paediatric inhaled medications: Children use (in most cases) inhalation devices initially developed for adults! 5

Target Identification HTS Hit-to-Lead Lead Generation Lead Optimisation Candidate Pre-clinical Phase I-III NDA FDA review Approval Role for lung imaging (pre-clinical species or human tissue): I. Understanding disease progression II. Biomarker identification Role for lung imaging (pre-clinical species): I. Deposition in animals II. Monitoring disease progression III. Monitoring drug efficacy IV. Biomarker identification But: Scintigraphic deposition studies in children are in most countries seen as not appropriate! Role for lung imaging (clinical studies): I. Deposition in man II. Monitoring disease progression III. Monitoring drug efficacy IV. Biomarker validation and monitoring

Ethical dilemma: Efficient, safe and wellstudied medications meeting the special needs of children! Avoiding any risks during tests with children & minimizing the number of necessary clinical trials! Performing realistic in vitro tests already at an early stage of development via artificial casts of children s upper airway geometry and simulating their specific inhalation profile! => Tools: Clinical handling studies & MRI/CT scans of children 7

Contents: Motivation: Why modelling the paediatric dose to lung via in-vitro and in-silico methods? Experiences with adult models Exploring the paediatric anatomy and physiology Why to use paediatric models at an early stage of product development? Challenges in the lab Summary & Outlook 8

First step: Can adult upper airway models really mimick the in-vivo conditions? to pressurized air (+humidifier) conditioning cabinet to electronic lung to flow pump Mouth-throat model + Respimat NGI 9

Comparison to published scintigraphic in-vivo data reveiled good agreement to our in-vitro results! Spiriva HandiHaler: Lung dose for patients with very severe COPD In-vitro data: 15.05% Clinical data: 14% (Throat: 34.36%) (Throat: 29%) Respimat: Lung dose for patients with very severe COPD In-vitro data: 65.26% Clinical data: 53% (Throat: 34.81%) (Throat 45%) pmdi: Lung dose for patients with very severe COPD In-vitro data: 24.2% Clinical data: 21% (Throat: 56.62%) (Throat 56%) *Brand et al. (2007) 10

Flow rate accelerates Good agreement between in-vitro and in-silico data using an adult upper airway model. Adult model CFD-Code: Fluent 6.3 Turbulence model: k-ώ SST Contourplot of Flow Velocity Particle Velocity 11

Contents: Motivation: Why modelling the paediatric dose to lung via in-vitro and in-silico methods? Experiences with adult models Exploring the paediatric anatomy and physiology Why to use paediatric models at an early stage of product development? Challenges in the lab Summary & Outlook 12

EU Pediatric Regulation 1901/2006: Regulation came into force on 26 January, 2007 Pediatric Committee of the EMA (PDCO) established July 2007 for the approval of new pharmaceuticals (or new indications / dosage forms) clinical studies for children have to be conducted (deferrals / waivers possible)! Rewarding: 6 months SPC extension, PUMA (orphan drug: 2 years additional market exclusivity 10 + 2) 13

Deposition data from adults cannot be extrapolated to children, because children have different Breathing pattern lower inspiratory flow and volume shorter inhalation time < 2 yrs preferred nose breathing high breathing frequency depending on kind of disease, mood (relaxed/quiet vs anxious/stressed), body size and device used Anatomy different upper airway geometry (not only smaller) large ratio of tongue to oral cavity the respiratory zones (terminal bronchioles, alveoles) are growing till the age of eight years in number, later in size glottis and subglottis aren t fully developed at birth Behaviour < 3 yrs are often unable to adopt a prescribed breathing pattern motor skills are impaired a lot of children become upset and distressed => a short application time is recommended! Subglottis of a 2-month (left side) and a 10-month old child (right side). 14

Workflow: med. data DICOM Patient geometry data idealized model segmentation Inhaler? simplification, approximation, im-/export CAD Pharmaceutical ingredient Inhalation flow profiles STL-output rapid prototyping / CNC manufact. / Laboratory bench testing Mesh generation / Computational Fluid Dynamics (CFD)

Analyzing 78 MRI and CT scans of children aged 3 days to five years reveil a high anatomical diversity! And clinical handling studies of inhalation profiles do also! 16

Analyzing 78 MRI and CT scans of children aged 3 days to five years reveil a high anatomical diversity! a) b) 4 2 3 6 5 8 3 7 17

Realistic models of children s extrathoracic airways for in vitro inhaler testing. V=0.0398 L V=0.0245 L V=0.0148 L V=0.0112 L for comparison: adult model V=0.079 L Age 4-5 yrs Age 3-4 yrs Age 2-3 yrs Age 1-2 yrs 18

Contents: Motivation: Why modelling the paediatric dose to lung via in-vitro and in-silico methods? Experiences with adult models Exploring the paediatric anatomy and physiology Why to use paediatric models at an early stage of product development? Challenges in the lab Summary & Outlook 19

In the wonderland of spacer devices which one to choose? 20

% of label claim Different types of spacers / valved holding chambers lead to different throat deposition and dose to lung. 100 80 60 40 20 DTL 51% Throat DTL 22% DTL 33% Throat DTL 33% Throat Results using throat model and flow profiles a) single breath Respimat and b) 5 breaths with spacers of a 5 year-old child. 0 RH=50% Respimat Funhaler Vortex Aerochamber Plus Mouthpiece Respimat Spacer Throat DTL Can Pediatric Throat Models and Air Flow Profiles Improve Our Dose Finding Stratety? Herbert Wachtel,, Jorg Breitkreutz, Peter Langguth RDD 2010, Vol 1 (2010): pp 195-204 21

% drug deposition related to the label claim Deposition results using our realistic child s model (age 2-3 yrs) show good correlation with in vivo data! Age 2-3 yrs 100 90 80 70 60 50 40 30 20 10 0 In vivo (Wildhaber et al.*) In vitro (our model) Lung deposition Actuator VHC * J.H. Wildhaber et al. (1999), The Journal of Pediatrics, Vol. 135, No. 1, pp 28-33 Link to scintigraphic deposition data of Wildhaber et al. (1999) in six 2 year-old children using the same pressurized Metered Dose Inhaler (albuterol) and valved holding chamber (AeroChamberPlus with facemask). 22

Computational Fluid Dynamics (CFD) studies provide an impressive insight on the different flow fields. 10 Particle velocity [m/s] Inhalation flow rate: 20 L/min Child s model 0 Adult model CFD-Code: Fluent 6.3 k ω SST turbulence model The child s model shows. - higher particle velocities (narrow laryngeal region & trachea) - more vortices (jolted pharyngeal region) resulting in a higher amount of particle deposition! 23

Particle throat deposition [%] related to the delivered dose Good agreement between in-vitro and in-silico results using the paediatric anatomical model. 100 90 80 70 Experiment Child (5yrs) Simulation 60 50 40 30 20 10 0 5 l/min 10 l/min 20 l/min 30 l/min CFD-Code: Fluent 6.3 Turbulence model: k-ώ SST Experimental data with constant flow rates as well as CFD simulations were conducted using the Respimat Soft Mist Inhaler. 24

Contents: Motivation: Why modelling the paediatric dose to lung via in-vitro and in-silico methods? Experiences with adult models Exploring the paediatric anatomy and physiology Why to use paediatric models at an early stage of product development? Challenges in the lab Summary & Outlook 25

The analytical challenge where is the drug in these numerous parts? 26

Further issues: Finding appropriate CT / MRI scans of children, especially infants Harmonized paediatric upper airway models for better comparison reasons? => Under discussion in the ISAM paediatric networking group (which age classes, extension to the first bifurcations of the lung?) Improved turbulence models for better resolution of the complex flow in the human upper airways Standardized inhalation profiles? 27

Summary In vitro and in silico modelling of inhaler performance in children via paediatric throat models based on CT and MRI scans help: to understand the special needs of infants, toddlers and pre-school children for a successful pulmonary treatment. Thank you for your attention! Questions? to design appropriate inhalation devices for children s special anatomy/physiology & enhanced compliance. to support dose finding strategies for inhaled medications while preventing children being exposed to investigational medicinal products (plus saving time & costs). 28

Acknowledgement - Dr. Herbert Wachtel, Boehringer Ingelheim - Andree Jung, Sarina Linz and Matthias Metzger, Boehringer Ingelheim - Dr. Ralf Kröger, Ansys Germany - Professor Dr. Peter Langguth, Johannes Gutenberg-University, Mainz - Professor Dr. Jörg Breitkreutz, Heinrich Heine-University, Düsseldorf 29